Researchers and clinicians have proposed that an infant shows individual characteristics of behavior or temperament that will be relatively stable. Thomas, et al. (1968) have concluded that a particular cluster of temperament characteristics, including high activity level, negative mood, lack of adaptability, and other variables, defines a baby as a "difficult infant." The difficult infant, who makes his caregivers' work more demanding and less satisfying, was found by Thomas, it al. to be "high risk" i.e., more likely than other children to develop behavioral problems in early childhood. While it is provocative, this work has methodological flaws. It needs to be replicated and extended. The present study will attempt to do so. First, a well-validated parent-report instrument measuring degree of difficultness in 4-6 month infants has been developed. Second, using this questionnaire, samples of difficult, average, and easy babies will be drawn from all segments of the population and studied in comparison with one another. Since one result of the difficult infant syndrome, as well as a major determinant of later problems may be less-than-optimal mother-infant interaction, this relationship will be given special attention, both in naturalistic and structured settings, and in large- and fine-grained analysis. Third, the children's characteristics and developmental progress will be monitored, for some children until age 2, to see whether difficultness and poor mother-infant relationship do indeed predict later problems.